Paxton Fitzpatrick

About me

I build computational models of human learning and memory, and create software tools for research and education.

I’m currently a Ph.D. student in Cognitive Neuroscience at Dartmouth College, working with Dr. Jeremy Manning and Dr. Luke Chang.
I’ve always been fascinated by the question of what makes people who they are.
X Growing up, I spent a lot of time meeting new people in new places. I lived in seven different cities throughout the northeastern and midwestern US before settling in a small town outside Harrisburg, Pennsylvania, where I continued to travel frequently for tennis, music, and quiz bowl competitions. I constantly encountered people with vastly different personalities and beliefs, and was always curious how those traits arose from each individual’s unique set of life experiences.
X When I came to Dartmouth College as an undergraduate in 2015 and discovered there was an entire field devoted to these questions, I quickly became excited about neuroscience research. I began working for the Dartmouth Brain Imaging Center, collecting and processing fMRI data for a variety of imaging studies, as well as Bregman Media Labs, where I worked with Dr. Michael Casey to create Exnectome, the first EEG sonification-based musical ensemble.
X I then joined the Contextual Dynamics Lab and discovered passions for computational memory research and open-source software development. Over time, I developed a love for mentoring, and in my junior year of college, began managing the CDL and volunteering as a tutor and TA for various neuroscience and computer science classes. I graduated in 2019 and wrote my honors thesis on a computational framework for modeling how episodic memories deform over time, and predicting neural activity from their semantic content.
X I continued to manage the CDL for two more years, designing naturalistic memory studies, building research software tools, TAing classes, and mentoring undergrads. Then, in 2021, I decided to stay at Dartmouth to pursue a Ph.D. in cognitive neuroscience, now working jointly in the CDL and the Computational Social Affective Neuroscience Lab. My current research is focused on leveraging the models of learning and memory I’ve developed in prior work to improve STEM accessibility and student outcomes by developing tools that allow instructors to easily and scalably personalize online courses.
  • Undergraduate, Dartmouth College
  • Research Assistant, Dartmouth Brain Imaging Center
  • Research Assistant, Contextual Dynamics Lab
  • Research Assistant, Bregman Media Labs
  • Laboratory & Research Manager, Contextual Dynamics Lab
  • Peer Tutor, Intro Computer Science
  • Teaching Assistant, Human Memory
  • Teaching Assistant & Guest Lecturer, Intro to Programming for Psychological Scientists
  • Teaching Assistant, Storytelling with Data
  • Teaching Assistant, Intro to Programming for Psychological Scientists
  • Ph.D. Student, Dartmouth College
  • Teaching Assistant, Laboratory in Experimental Psychology
Programming & web development
Python
JavaScript
Unix shell
LaTeX
SuperCollider
HTML
CSS
Sass/SCSS
Liquid
Jekyll
Development tools
Git/GitHub
Docker
SQLite/MySQL/SQLAlchemy
Moab-TORQUE
Conda
pytest
Selenium
Travis CI
GitHub Actions
Jupyter/IPython/ Colaboratory
Python packaging
Experimental design & data collection
jsPsych
psiTurk
Amazon Mechanical Turk
PsychoPy
Cedrus SuperLab
OpenBCI
OpenSesame
Norkross MorphX
Qualtrics
Data analysis & visualization
SciPy
NumPy
Pandas
NLTK
statsmodels
Scikit-learn
Nilearn
igraph
Graphia
Matplotlib
Seaborn
Plotly
Hypertools
FreeSurfer
BORIS
ANVIL
Non-technical
Adobe Illustrator
Adobe Photoshop
French
Scientific & expository writing
Public speaking
Organizational leadership

Research

My research uses language models to capture how our memories preserve, distort, and compress the external world, and to identify the neural mechanisms that underlie these transformations.

I’m particularly interested in how episodic memory compresses the temporal structure of an experience, and how that compression is warped over time.
X

Davos: The Python package smuggler

Fitzpatrick P. C., Manning J. R. (2022). Davos: The Python package smuggler. (under revision).

X

Fitness tracking reveals task-specific associations between memory, mental health, and exercise

Manning J. R., Notaro G. M., Chen E., Fitzpatrick P. C. (2022). Fitness tracking reveals task-specific associations between memory, mental health, and exercise. (under revision).

X

A geometric approach to modeling knowledge and learning from Khan Academy course videos

Fitzpatrick P. C., Heusser A. C., Manning J. R. (2022). A geometric approach to modeling knowledge and learning from Khan Academy course videos. Context and Episodic Memory Symposium. Philadelphia, PA.

X

Cognitive tasks as a diagnostic tool for mental health

Jain S., Schreder N., Fitzpatrick P. C., Ziman K., Manning J. R. (2021). Cognitive tasks as a diagnostic tool for mental health. Trends in Psychology Summit. Cambridge, MA.

X

Capturing the geometric and neural structures of experiences and memories

Fitzpatrick P. C. (2022). Capturing the geometric and neural structures of experiences and memories. Dartmouth College. Hanover, NH.

X

Geometric models reveal behavioral and neural signatures of transforming experiences into memories

Heusser A. C.†, Fitzpatrick P. C.†, Manning J. R. (2021). Geometric models reveal behavioural and neural signatures of transforming experiences into memories. Nature Human Behaviour. doi:10.1038/s41562-021-01051-6.

†denotes equal contribution

X

Docker for scientific research

Fitzpatrick P. C. (2021). Docker for scientific research. Dartmouth College. Hanover, NH.

X

Web-based behavioral experiments for online data collection

Fitzpatrick P. C. (2020). Web-based behavioral experiments for online data collection. EPSCoR Attention Consortium meeting. Virtual.

X

Exploring the evolving geometric structure of experiences and memories

Fitzpatrick P. C., Heusser A. C., Manning J. R. (2019). Exploring the evolving geometric structure of experiences and memories. Society for Neuroscience Annual Meeting. Chicago, IL.

X

Mapping between naturalistic experience and verbal recall

Fitzpatrick P. C., Heusser A. C., Manning J. R. (2018). Mapping between naturalistic experience and verbal recall. Society for Neuroscience Annual Meeting. San Diego, CA.

X

Capturing the geometric structure of our experiences and how we remember them

Heusser A. C., Fitzpatrick P. C., Manning J. R. (2018). Capturing the geometric structure of our experiences and how we remember them. Conference on Cognitive Computational Neuroscience. Philadelphia, PA.

X

The utility of speech-to-text software for transcription of verbal response data

Fitzpatrick P. C., Ziman K., Heusser A. C., Field C. E., Manning J. R. (2018). The utility of speech-to-text software for transcription of verbal response data. Wetterhahn Science Symposium. Hanover, NH.

X

Adaptive free recall: Enhancing (or diminishing) memory

Lee M., Chacko R., Whitaker E., Fitzpatrick P. C., Field C. E., Ziman K., Bollinger B., Heusser A. C., Manning J. R. (2018). Adaptive free recall: Enhancing (or diminishing) memory. Wetterhahn Science Symposium. Hanover, NH.

X

Is automatic speech-to-text transcription ready for use in psychological experiments?

Ziman K., Heusser A. C., Fitzpatrick P. C., Field C. E., Manning J. R. (2018). Is automatic speech-to-text transcription ready for use in psychological experiments?. Behavior Research Methods, 1-9.

X

Quail: a Python toolbox for analyzing and plotting free recall data

Heusser A. C., Fitzpatrick P. C., Field C. E., Ziman K., Manning J. R. (2017). Quail: a Python toolbox for analyzing and plotting free recall data. The Journal of Open Source Software, 2(18): 424.

X

Davos: The Python package smuggler

Fitzpatrick P. C., Manning J. R. (2022). Davos: The Python package smuggler. (under revision).

X

Fitness tracking reveals task-specific associations between memory, mental health, and exercise

Manning J. R., Notaro G. M., Chen E., Fitzpatrick P. C. (2022). Fitness tracking reveals task-specific associations between memory, mental health, and exercise. (under revision).

X

Geometric models reveal behavioral and neural signatures of transforming experiences into memories

Heusser A. C.†, Fitzpatrick P. C.†, Manning J. R. (2021). Geometric models reveal behavioural and neural signatures of transforming experiences into memories. Nature Human Behaviour. doi:10.1038/s41562-021-01051-6.

†denotes equal contribution

X

Is automatic speech-to-text transcription ready for use in psychological experiments?

Ziman K., Heusser A. C., Fitzpatrick P. C., Field C. E., Manning J. R. (2018). Is automatic speech-to-text transcription ready for use in psychological experiments?. Behavior Research Methods, 1-9.

X

Quail: a Python toolbox for analyzing and plotting free recall data

Heusser A. C., Fitzpatrick P. C., Field C. E., Ziman K., Manning J. R. (2017). Quail: a Python toolbox for analyzing and plotting free recall data. The Journal of Open Source Software, 2(18): 424.

X

Capturing the geometric and neural structures of experiences and memories

Fitzpatrick P. C. (2022). Capturing the geometric and neural structures of experiences and memories. Dartmouth College. Hanover, NH.

X

Docker for scientific research

Fitzpatrick P. C. (2021). Docker for scientific research. Dartmouth College. Hanover, NH.

X

Web-based behavioral experiments for online data collection

Fitzpatrick P. C. (2020). Web-based behavioral experiments for online data collection. EPSCoR Attention Consortium meeting. Virtual.

X

A geometric approach to modeling knowledge and learning from Khan Academy course videos

Fitzpatrick P. C., Heusser A. C., Manning J. R. (2022). A geometric approach to modeling knowledge and learning from Khan Academy course videos. Context and Episodic Memory Symposium. Philadelphia, PA.

X

Cognitive tasks as a diagnostic tool for mental health

Jain S., Schreder N., Fitzpatrick P. C., Ziman K., Manning J. R. (2021). Cognitive tasks as a diagnostic tool for mental health. Trends in Psychology Summit. Cambridge, MA.

X

Exploring the evolving geometric structure of experiences and memories

Fitzpatrick P. C., Heusser A. C., Manning J. R. (2019). Exploring the evolving geometric structure of experiences and memories. Society for Neuroscience Annual Meeting. Chicago, IL.

X

Mapping between naturalistic experience and verbal recall

Fitzpatrick P. C., Heusser A. C., Manning J. R. (2018). Mapping between naturalistic experience and verbal recall. Society for Neuroscience Annual Meeting. San Diego, CA.

X

Capturing the geometric structure of our experiences and how we remember them

Heusser A. C., Fitzpatrick P. C., Manning J. R. (2018). Capturing the geometric structure of our experiences and how we remember them. Conference on Cognitive Computational Neuroscience. Philadelphia, PA.

X

The utility of speech-to-text software for transcription of verbal response data

Fitzpatrick P. C., Ziman K., Heusser A. C., Field C. E., Manning J. R. (2018). The utility of speech-to-text software for transcription of verbal response data. Wetterhahn Science Symposium. Hanover, NH.

X

Adaptive free recall: Enhancing (or diminishing) memory

Lee M., Chacko R., Whitaker E., Fitzpatrick P. C., Field C. E., Ziman K., Bollinger B., Heusser A. C., Manning J. R. (2018). Adaptive free recall: Enhancing (or diminishing) memory. Wetterhahn Science Symposium. Hanover, NH.

Software

I also regularly develop, maintain, and contribute to a number of open-source software projects.

davos

Import Python packages even if they aren’t installed. Enables the “smuggle” statement: a drop-in replacement for import that handles missing packages on the fly. Add “onion comments” alongside smuggle statements specify package versions and additional options. Can be used to turn Jupyter or Colab notebooks into self-contained, reproducible Python environments that manage dependencies at runtime.

docker-tutorial

Walkthroughs and template code for running experiments and analyzing data from within Docker containers. Pre-built images are available on Docker Hub.

particle-image

Animate a particlized image in vanilla JavaScript. Turn an image from the web into an animated swarm of interactive particles. Set parameters to control particle color, size, density, speed, hover/click/touch interactions, and more.

psiturk-experiment-template

A template behavioral experiment ready to be deployed locally or online via Amazon Mechanical Turk. Implemented using the psiTurk platform and jsPsych library, and isolated within four networked Docker containers: a Debian 9 container to house the experiment code and psiTurk server, an nginx server for load balancing, a MySQL database for storing data, and Adminer for inspecting and downloading data.

CDL-docker-stacks

A collection of optimized, extensible, hierarchically built Docker images for common neuro/data science tasks. Pre-built images are available on Docker Hub in Python 3.6, 3.7, and 3.8 variants.

cluster-tools-dartmouth

A Python toolbox for remotely interacting with Dartmouth’s Discovery HPC cluster. Automatically generates PBS scripts, submits jobs to the scheduler, monitors progress, and compiles results when finished. Can easily be configured to work with any Moab/TORQUE system.

gittracker

A CLI app for tracking the states of all your local Git repositories in one place. Run a single command from any directory to display git status-like information for each repo gittracker is configured to track. Supports arbitrarily nested submodules, multiple verbosity levels, and automatic discovery of local repos.

quail

A Python toolbox for processing, analyzing, and visualizing free recall data. Provides a common interface for working with data from both list-learning and naturalistic memory experiments. Integrates with the Google Cloud Speech-to-Text API for rapid, on-the-fly audio transcription.

autoFR

A verbal free recall experiment that incorporates automated speech decoding. Uses quail to automatically obtain recall transcripts, onset/offset times, IRTs, confidence scores, and other metadata. Provides a custom jsPsych plugin for collecting and saving verbal recall data.

hypertools # lead maintainer

A Python package for visualizing and manipulating high-dimensional data. Transform, align, normalize, interpolate, cluster, reduce, and plot numeric or text data with a single function call or in individual steps. Save full analysis pipelines including data and trained models as DataGeometry objects for later reuse. Designed to be fully customizable with reasonable defaults.

umap-learn # core maintainer

Python implementation of Uniform Manifold Approximation and Projection. A general-purpose non-linear dimensionality reduction algorithm based on finding a low-dimensional projection of the data that best preserves its fuzzy topological structure. The Python package also implements supervised, semi-supervised, aligned, and parametric UMAP variants, an inverse transform, densMAP, and a plotting library.

timecorr # core contributor

A Python toolbox for exploring higher-order structure in multivariate timeseries data. Iteratively computes dynamic correlations and reduces dimensionality to approximate higher-order dynamic correlations in a computationally tractable way. Also supports computing various dynamic graph theoretic measures.

SuperEEG # core contributor

Infer activity throughout the brain from a small(ish) number of electrodes using Gaussian process regression

CDL-bibliography # core contributor

A shared bibtex file containing references for ~6,000 psychology, neuroscience, math, and machine learning papers.

lab-manual # core maintainer

Lab manual and associated source code for the Contextual Dynamics Lab at Dartmouth College.

…and of course this website!

Get in touch

Contact info

Paxton.C.Fitzpatrick.GR@Dartmouth.edu

@paxt0n4

416 Moore Hall, Dartmouth College
Hanover, NH 03755

Where am I?